• 沒有找到結果。

Combined photolysis and catalytic ozonation of dimethyl phthalate in a high-gravity rotating packed bed

N/A
N/A
Protected

Academic year: 2021

Share "Combined photolysis and catalytic ozonation of dimethyl phthalate in a high-gravity rotating packed bed"

Copied!
7
0
0

加載中.... (立即查看全文)

全文

(1)

Contents lists available atScienceDirect

Journal of Hazardous Materials

j o u r n a l h o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / j h a z m a t

Combined photolysis and catalytic ozonation of dimethyl phthalate in a

high-gravity rotating packed bed

Chia-Chi Chang

a

, Chun-Yu Chiu

b

, Ching-Yuan Chang

a,∗

, Chiung-Fen Chang

c

,

Yi-Hung Chen

d

, Dar-Ren Ji

a

, Yue-Hwa Yu

a

, Pen-Chi Chiang

a

aGraduate Institute of Environmental Engineering, National Taiwan University, Taipei 106, Taiwan bDepartment of Cosmetic Science and Application, Lan-Yang Institute of Technology, I-Lan 261, Taiwan cDepartment of Environmental Science and Engineering, Tunghai University, Taichung 407, Taiwan

dDepartment of Chemical and Material Engineering, National Kaohsiung University of Applied Science, Kaohsiung City 807, Taiwan

a r t i c l e i n f o

Article history: Received 4 January 2008

Received in revised form 19 March 2008 Accepted 19 March 2008

Available online 26 March 2008 Keywords:

Ozonation

High-gravity rotating packed bed (HGRPB) Photolysis

Dimethyl phthalate (DMP) Catalyst

a b s t r a c t

In this study, a high-gravity rotating packed bed (HGRPB) was used as a catalytic ozonation reactor to decompose dimethyl phthalate (DMP), an endocrine disrupting chemical commonly encountered. The HGRPB is an effective gas–liquid mixing equipment which can enhance the ozone mass transfer coeffi-cient. Platinum-containing catalyst (Pt/-Al2O3) of Dash 220N and ultra violet (UV) lamp were combined

in the high-gravity ozonation (HG-OZ) system to enhance the self-decomposition of molecular ozone in liquid to form highly reactive radical species. Different combinations of HG-OZ with Dash 220N and UV for the degradation of DMP were tested. These include HG-OZ, HG catalytic OZ (HG-Pt-OZ), HG photolysis OZ (HG-UV-OZ) and HG-UV-Pt-OZ. The result indicated that all the above four ozonation pro-cesses result in significant decomposition of DMP and mineralization of total organic carbon (TOC) at the applied ozone dosage per volume of liquid sample of 1.2 g L−1. The UV and Pt/␥-Al2O3combined in

HG-OZ can enhance the TOC mineralization efficiency (TOC) to 56% (via HG-UV-OZ) and 57% (via

HG-Pt-OZ), respectively, while only 45% with ozone only. The process of HG-UV-Pt-OZ offers the highestTOCof

about 68%.

© 2008 Elsevier B.V. All rights reserved.

1. Introduction

Endocrine disrupting chemicals (EDCs, also called environ-mental hormones) have been spread over the environment and influenced the generative function of some species on the earth. Most of EDCs are persistent organic pollutants (POPs) which exhibit chronic toxicity, bioaccumulation and are difficult to be decom-posed by organisms in the environment. Among the EDCs, phthalate acid esters (PAEs) are chemical compounds most widely used in the plastic industrial production. The major applications of PAEs are for the use as plasticizers for polyvinyl chloride (PVC) resins, adhe-sives and cellulose film coating (counting about 85% of the whole production). The short-chained ester such as dimethyl phthalate (DMP) is among the most frequently identified PAEs in diversi-fied environmental samples including upper layer of marine water, surface waters and sediments. Esters are found to accumulate in the environment and to be toxic to a variety of aquatic organisms, which are at the base of the food chain in both marine and

sur-∗ Corresponding author. Tel.: +886 2 2363 8994; fax: +886 2 2363 8994. E-mail address:[email protected](C.-Y. Chang).

face water environments. Destruction of these compounds requires the application of processes having high decomposition abilities such as those being able to provide highly reactive hydroxyl radi-cals. Recently, researches showed that phthalate esters in secondary effluent and surface water can be decomposed via O3/UV processes [1,2].

Ozonation processes have been widely applied to water and wastewater treatment. In aqueous solution, ozone acts on various compounds by direct ozonation and radical reaction. Molecular ozone is a powerful oxidant agent which reacts selectively with organic compounds in water. The organic compounds with molec-ular structure consisting of multiple bonds (for example C C, C C O R and C C X) or atom carrying a negative charge (such as N, P, O, S and nucleophilic carbon) are highly reactive with ozone [3]. The ozonation system should be able to form highly reactive radical species from the decomposition of molecular ozone in water so as to destruct the POPs via free radical chain reaction. Alternative ozonation processes using catalysts of transition metals also have been applied for the degradation of organics. Catalytic ozonation can be classified according to (1) activation of ozone via metal ions in solution and (2) heterogeneous catalytic ozonation in the pres-ence of metal oxides or metals on supports[4]. The use of metal 0304-3894/$ – see front matter © 2008 Elsevier B.V. All rights reserved.

(2)

288 C.-C. Chang et al. / Journal of Hazardous Materials 161 (2009) 287–293

Nomenclature

a specific area of gas–liquid interface per unit volume of contactor or packed bed (m2m−3)

AOPs advanced oxidation processes

BET Brunauer–Emmett–Teller surface adsorption

C concentration of DMP (mg L−1)

C0 initial concentration of DMP (mg L−1)

CAG,in concentration of feed O3(mg L−1) CSTR completely stirred tank reactor

DMP dimethyl phthalate

EDC endocrine disrupting chemical

HGRPB high-gravity rotating packed bed

HG-OZ HGRPB-ozonation

HG-Pt-OZ Pt catalytic HG-OZ HG-UV-OZ ultra violet HG-OZ HG-UV-Pt-OZ Pt catalytic HG-UV-OZ HRT hydraulic retention time (s) k0

LA physical liquid-phase mass transfer coefficient of ozone (m s−1)

mA,in applied dosage of ozone per volume of liquid sample (mg L−1)

mS mass of catalyst (g)

Nr rotating speed of HGRPB (rpm)

PAEs phthalate acid esters POPs persistent organic pollutants

QG flow rate of feed O3(L min−1)

QLR flow rate of recycled liquid (L min−1)

RB5 Reactive black 5

RB19 Reactive blue 19

t reaction time (min)

T temperature (◦C)

TOC total organic carbon

UV ultra violet

VL sample volume (L)

Greek letters

εB void fraction of packed bed TOC mineralization efficiency of TOC

ions as the homogeneous catalyst has no mass transfer resistance between the reactants and catalyst. However, the metal ion remains in the solution and is not easy to be recovered or removed. Solid cat-alysts which are packed in the packed bed are able to be recycled and regenerated. The catalysts with metals such as Pt, Ru, Rh, Pd and Ag were reported to be effective for the decomposition of organic compounds in gaseous as well as aqueous-phase ozonation. The order of catalytic activity of metals supported on Al2O3was found to be: Pt > Pd > Ag > Ru∼ Rh ∼ Ir > Ni > Cd > Mn > Fe > Cu > Zn ∼ Zr[5]. In this study, platinum-containing catalyst, Dash 220N, was chosen as the catalyst to produce the hydroxyl radicals from aqueous ozone. Ozonation is limited by its selective reactivity and low gas–liquid mass transfer rate. The high-gravity rotating packed bed (HGRPB, HG or Higee) was used in this study as a gas–liquid contactor to enhance the mass transfer rate between the phases. The values of volumetric gas–liquid mass transfer coefficients achievable in HGRPB are one or two orders of magnitude higher than those in con-ventional packed beds[6]. Recent studies[7–9]showed that HGRPB has some advantageous characteristics such as (1) fast renewable rate on the surface of packed materials, (2) high gas–liquid mass transfer coefficients, (3) short time to reach steady state, (4) low overflow rate, (5) short hydraulic retention time (HRT) and (6) thin liquid film.

HGRPB is also applicable to absorption, adsorption, desorption, distillation, polymer devolatilization, bio-oxidation, reactive crys-tallization, stripping, extraction and other separation processes [10]. Recently, HGRPB also has been used for micro-mixing and emulsion[11,12].

Previous studies point out that HGRPB has a higher mass transfer rate in gas–liquid system than conventional gas–liquid contac-tor. Thus, the ozonation process can be combined with HGRPB to enhance the ozone mass transfer from gas phase to liquid phase. Lin and Liu[13]found that the centrifugal force can facilitate the decol-orization efficiency of dyestuff Reactive Blue 19 (RB19) because the mass transfer rate of ozone is enhanced via increasing the inter-facial area and decreasing the resistance of ozone transfer. Chen et al. [14] used RPB as well as completely stirred tank reactor (CSTR) as ozone gas–liquid contactor to treat CI Reactive Black 5 (RB5). The results indicated that the mass transfer coefficient of RPB (k0

LAa = 0.0975 s−1) is significantly higher than that of CSTR (k0

LAa = 0.0248 s−1).

In this research, the HGRPB was applied to the catalytic ozonation processes using Pt/␥-Al2O3 to decompose the organic compound of DMP in aqueous solution. System performances for the degradation of DMP via various advanced oxidation processes (AOPs) using HGRPB were examined and compared. These HGRPB assisted AOPs include HGRPB-ozonation (OZ), Pt catalytic HG-OZ (HG-Pt-HG-OZ), ultra violet HG-HG-OZ (HG-UV-HG-OZ) and HG-UV-Pt-HG-OZ. The results can provide insight information for the feasibility eval-uation of applying a HGRPB to the said AOPs.

2. Experimental

2.1. Chemicals

DMP with chemical formula of C10H10O4, molecular weight of 194.19 and purity >99% was purchased from Hayashi Pure Chem-ical Industries Ltd. (Osaka, Japan). The water solubility of DMP is 4000 mg L−1 at 25◦C (1 atm). The DMP was subjected to DI water prior to the use with the concentration (CDMP) of 100 mg L−1. The initial pH of the DMP solution is about 7.5. The potassium iodine (KI), sodium chloride (NaCl) and acetonitrile (CH3CN) used were of analytical grade (J.T. Baker, Phillipsburg, NJ, USA). Ozone was decomposed to form radicals via a commercial Pt/␥-Al2O3 catalyst Dash 220N (N.E. Chemical Co., Japan), which is a platinum-containing catalyst. Characteristic of the Dash 220N is shown in Table 1 [15]. Dash 220N is a spherical catalyst 3–4 mm in diameter. The bulk density of the Dash 220N is 0.77 g cm−3. The content of platinum in the alumina sphere is about 0.23 wt.%.

2.2. Instrumentation

The HGRPB system (Fig. 1) was carried out via semi-batch opera-tion. The system has two major parts, namely RPB and photoreactor. The RPB consists of a packing chamber, a rotor and a stationary housing. The DMP solution flows outward from the inner edge of

Table 1

Characterization of Pt/␥-Al2O3DASH 220N catalyst[13] Average particle diameter, dP 4 mm

Total BET surface area, AB 166.62 m2g−1

Total pore volume, Vt 0.4392 cm3g−1

Average pore diameter, dPr 76 ˚A

Bulk particle density,P 0.77 g cm−3

Ture density,s(He displacement method) 3.22 g cm−3

Porosity,εP 0.59

Pt content in catalyst 0.23 wt.%

(3)

Fig. 1. Schematic diagram of HGRPB system.

the packing chamber subjected to the centrifugal force. Gaseous ozone flows inward countercurrently from the outer edge of the packed bed by the pressure-driving force. The experiments were performed with the volume (VL) of DMP solution of 1 L and total reaction time (t) of 60 min. To identify the catalytic ability of plat-inum in ozonation system, two packing materials of Dash 220N and spherical glass beads were used and tested in the packed bed. The diameter of glass beads is about the same as that of Dash 220N. The RPB is 2 cm in height and 5.9 cm in diameter. In the experi-ments, the RPB was operated with rotating speed (Nr) from 600 to 1800 rpm, which provided gravitational force of 23.76–213.91 g (232.92–2096.31 m s−2). The DMP solution was fed to RPB by a peri-staltic pump at 462 mL min−1recycle rate (QLR) and contacted with gaseous ozone and catalysts in the bed.

The photoreactor is made of Pyrex glass with the dimensions of diameter and height of 5 and 30 cm, respectively. A UV lamp (TUV-16W, Phillps, Tokyo, Japan) was placed at the center of the UV reactor and shielded by a quartz jacket. The UV irradiation was introduced to enhance the decomposition of ozone to form the OH•radical. All the experiments were controlled at 25◦C with water jacket around the reactor. The DMP solution was pumped and recycled continuously through a closed loop connected with all sensors of pH, oxidation reduction potential (ORP) and dissolved ozone (DO3) monitors. Liquid samples were taken from a sampling port for chemical analyses.

The ozone was generated from dried oxygen with purity of 99.99% via corona discharge using an ozone generator (Model LAB2B, Ozonia, Duebendorf, Switzerland). The ozone inlet flow rate (QG) was controlled via a mass flow controller (MFC, Model 5850E, Brooks, Hatfield, PA, USA) at 1 L min−1. The inlet and outlet gaseous ozone concentrations were measured via a spectrophotometer (UV mini-1240, Shimadzu, Kyoto, Japan) with the absorbance of ozone measured in a 2 mm flow-through quartz cell at the wave-length 258 nm. An extinction coefficient of 3000 M−1s−1was used to convert absorbances into concentration units [16]. The inlet ozone concentration (CAG,in) was controlled at 25 mg L−1. Ozone

concentration in water was determined by DO3 meter (Model 3600, Orbisphere, Trasadingen, Switzerland). The UV–vis irradia-tion intensity was measured by diffracirradia-tion grating spectrometer (Model EPP 2000, StellarNet, Oldsmar, FL, USA).

The analyses of DMP solutions were performed using the high performance liquid chromatography (HPLC, Model 500, Viscotek, Houston, TX, USA) with 250 mm× 4.6 mm C18 column (LC-18, Supelco, Bellefonte, PA, USA) to separate the DMP and by-products of ozonation. The wavelength of UV/visible detector (Model 1706, Bio-Rad, Hercules, CA, USA) was set at 220 nm. Effluent is com-posed of CH3CN and DI water (CH3CN/DI water = 50/50) with flow rate controlled at 1.0 mL min−1. Total organic carbon (TOC) was ana-lyzed by TOC analyzer (Model 1010, O.I. analytical, College Station, TX, USA).

3. Results and discussion

3.1. Hydraulic test

Heterogeneous catalytic ozonation of DMP in HGRPB is a gas–liquid–solid reaction for which mass transfer and chemical reaction may play important roles. The mass transfer is depen-dent on the mixing of liquid, gaseous ozone and solid catalyst. The HGRPB is an efficient mixing apparatus with its high rotat-ing speed. The HRT of HGRPB was determined by observrotat-ing the variation of conductivity of tracer-containing effluent. The tracer (NaCl) was injected in the influent in the entrance of HGRPB.Fig. 2 shows the variations of HRT with Nrand QLR. At low water feeding rate (QLR= 188 mL min−1) and low rotating speed (Nr= 300 rpm), the HRT was about 10 s. The HRT reduced to 8 s as Nrincreased to 600 rpm for the case with QLR= 188 mL min−1. At the same QLR, the HRTs were about the same for the cases with Nrin the range of 600 to 1800 rpm. The HRTs were about 8, 6, 5, 4 and 4 at QLR of 188, 463, 545, 750 and 857 mL min−1. The result showed that at the same Nr, the HRT was strongly dependent on QLR. High Nr leads to high centrifugal force assisting the fluid spraying outward

(4)

290 C.-C. Chang et al. / Journal of Hazardous Materials 161 (2009) 287–293

Fig. 2. (a) Hydrolytic retention time (HRT) vs. rotating speed (Nr), (b) HRT vs. liquid flow rate (QLR).

through the packed bed. Liquid volume in HGRPB is limited by the voidage of the packed bed (εB). The total volume of the packed bed is 213.88 cm3with the volume of packed catalyst of 72.05 cm3and stainless wire of 18.27 cm3. This gives the free space of the packed bed of 123.56 cm3. Thus, theεBof HGRPB in this experiment was 57.77%. Theoretically, the HRTs without rotating would be 40, 16, 14, 10 and 9 s with QLRof 188, 462, 545, 750 and 875 mL min−1. Under high Nr, the liquid is speeded up via the centrifugal force, resulting in a significant decrease of HRT down to 4–10 s. However, the high

Nrof HGRPB can well mix the liquid with gaseous oxidant in a short time.

3.2. Photolysis and ozonation of DMP

The decomposition of DMP in HGRPB systems of HG, HG-UV, HG-OZ and HG-UV-OZ at QG= 1 L min−1, QLR= 100 mL min−1 and

Nr= 1200 rpm is shown inFig. 3. In this part of experiment, the HGRPB was packed with glass spheres 4 mm in diameter. For the cases without ozone, nitrogen gas was introduced to maintain the same hydraulic conditions. DMP was not decomposed using HG and HG-UV. For the ozonation, the applied ozone dosage (mA,in) and decomposition efficiency of DMP (DMP) are defined as follows:

mA,in= QG× CAG,in× t × VL−1 (1)

DMP=(CDMP0− CDMP)

CDMP0 (2)

where QG, CAG,in and VL are gas flow rate, concentration of inlet ozone and volume of liquid sample, while CDMP0(or C0) and CDMP (or C) are concentrations of DMP at time t = 0 and t, respectively.

Fig. 3. Time variations of C/C0 for the decomposition of DMP in HGRPB sys-tems. Initial concentration of DMP (CDMP0 or C0) = 100 mg L−1, QG= 1 L min−1, QLR= 462 mL min−1, power of UV lamp = 16 W as noted in specification, VL= 1 L, Nr= 1200 rpm, CAG,in= 20 mg L−1, T = 25◦C. (): HG only; (): HG-photolysis (HG-UV); (♦): HG ozonation (HG-OZ); (): HG-UV-OZ. HGRPB: high-gravity rotating packed bed.

The values of DMP for HG-OZ process reached about 80% in 30 min reaction time (mA,in= 600 mg L−1) and 96% in 60 min (mA,in= 1200 mg L−1), respectively, revealing effective decomposi-tion of DMP via ozonadecomposi-tion.

The emission spectrum of UV lamp (denoted as UV raw) for the photolysis experiments is shown inFig. 4.Fig. 4a shows the

Fig. 4. Absorption spectrum of DMP. Light source: 16 W UV lamp; CDMP0= 100 mg L−1.

(5)

raw emission spectrum of the UV lamp. It indicated that the low-pressure UV lamp employed is a multi-wavelengths light source with three regions of UV-A, UV-B and UV-C. The irradiation inten-sities of the UV lamp in UV-A (315–400 nm), UV-B (280–315 nm) and UV-C (200–280 nm) are 3.99, 1.59 and 3.73 W m−2 counting the whole wavelength range of the respective regions.Fig. 4b and c are UV spectrum through DI water and DMP solution. The DMP has obvious UV absorption below 240 nm, while ozone absorbs UV light of short wavelengths in 200–300 nm with the maximum absorption coefficient at 253.7 nm. As shown inFig. 4a and c, the UV lamp energy emitted was absorbed by DMP solution in UV-C region, while there was no absorption in UV-A and UV-B regions.

The results ofFig. 3for the photolysis of DMP were obtained using the low-pressure UV lamp, yieldingDMPof 2%. The corre-sponding removal efficiency (or mineralization efficiency) of TOC (TOC= (CTOC0− CTOC)/CTOC0) was relatively low. Thus, DMP is hard to be destructed via UV irradiation only. As UV irradiation was com-bined with HG-OZ system (HG-UV-OZ process), the decomposition of DMP reached 85% in 30 min reaction time (mA,in= 600 mg L−1) and 99% in 60 min (mA,in= 1200 mg L−1).

In UV-OZ process for the decomposition of DMP, molecules absorbing UV irradiation of UV-C region include:

(1) molecular ozone – enhancing the decomposition of ozone in water and generating highly reactive hydroxyl radicals, (2) DMP molecule – absorbing UV-C via the aromatic ring, (3) intermediates from the decomposition of DMP – absorbing UV

and reacting with ozone and hydroxyl radicals to yield other by-products (such as phenols–quinines–aromatic acids, short-chain aliphatic acids and short-short-chain aliphatic aldehydes). The mechanism of UV-OZ may be described as follows. In aque-ous phase, the dissolved ozone absorbs UV radiation to product hydrogen peroxide (H2O2) via the following reactions[17]:

O3+ H2O−→O2h + H2O2 (3)

O3+ H2O−→O2h + 2OH• (4)

The second step of UV-OZ process is to form hydroxyl radical. The H2O2may be decomposed further to form hydroxyl radical via photolysis or ozonation as follows.

Photolysis:

H2O2−→2OHh • (5)

Ozone enhanced decomposition:

H2O2↔ HO2−+ H+ (6)

HO2−+ O3→ HO2• + O3•− (7)

O3−+ H+→ HO3→ OH+ O2. (8)

Comparing the reactions in the above pathway, the reaction rate of H2O2via photolysis is quite slow (Eq.(5)). Therefore, the decom-position of H2O2 via ozone to form hydroxyl radical is the main pathway (Eqs.(6)–(8)) in this experiment.

The initial pH value of sample solution was about 7.5. For the HG and HG-UV processes, the pH values were remained at about 7.5 because of nearly no decomposition of DMP. Nevertheless, the pH value of DMP solution decreased during the ozonation because of the formation of intermediates such as aromatic acids and short-chain aliphatic acids. In the HG-OZ process, the pH value decreased from 7.5 to 6.0 and 3.6 in 5 and 15 min, respectively. As for the HG-UV-OZ process, the pH value changed from 7.5 to 5.9 and 6.2 in 5

Fig. 5. Effects of mA,inon the decomposition of DMP for the HG-OZ process at various Nr. CDMP0(or C0) = 100 mg L−1, QG= 1 L min−1, QLR= 462 mL min−1, VL= 1 L, CAG,in= 20 mg L−1, T = 25C. (×), (♦), (), () and (): Nr= 600, 900, 1200, 1600 and 1800 rpm.

and 15 min, respectively. At later period of HG-UV-OZ, the pH value of solution increased slightly because of further decomposition of organic acids.

3.3. Effects of mA,inand Nron the ozonation of DMP

The decomposition of DMP via the HG-OZ process at various

Nr of HGRPB is shown in Fig. 5. The conditions of the pro-cess were: CDMP0= 100 mg L−1, QG= 1 L min−1, QLR= 462 mL min−1,

VL= 1 L and CAG,in= 25 mg L−1. Five rotating speeds 600, 900, 1200, 1600 and 1800 rpm were tested to evaluate the effect of Nr on

C/C0. The corresponding gravitational forces resulted were 23.76, 53.48, 95.07, 148.55 and 213.91 g (232.92, 524.08, 931.70, 1455.77 and 2096.31 m s−2), respectively. The result showed that under the experimental conditions, the values ofDMPincrease signifi-cantly with Nr. As Nr increased from 600 to 1800 rpm, theDMP at mAG,in= 1200 mg L−1 increased from 83 promote to 98%. High rotating speed can enhance the ozone gas–liquid mass transfer, thus resulting in a highDMP. Lin and Liu[13]also noted that an increase of Nrfor a HG-OZ process increases the decolorization of RB19 because the centrifugal force increases the interfacial area while decreases the resistance of ozone mass transfer via decreas-ing the thickness of liquid film and size of droplets. The results of Fig. 5further showed that as Nrincreases higher than 1200 rpm, the value ofDMPonly increases slightly, indicating the

minimiza-Fig. 6. C/C0vs. mA,induring ozonation of DMP in HGRPB. CDMP0(or C0) = 100 mg L−1, QG= 1 L min−1, QLR= 462 mL m−1, mass of Dash 220N (mS) = 220 g, power of UV lamp as indicated in specification = 16 W, VL= 1 L, Nr= 1200 rpm, CAG,in= 20 mg L−1, T = 25C. (♦): HG-OZ; (): HG catalytic ozonation (HG-Pt-OZ); (×): HG-UV-Pt-OZ.

(6)

292 C.-C. Chang et al. / Journal of Hazardous Materials 161 (2009) 287–293

Fig. 7. Scheme illustrating the possible mechanism for ozonation process in the

presence of metal-support (metal oxide) catalyst (based on). Me: metal oxide; AH: organics; ox: oxidation state; red: reduction state[2].

tion of mass transfer resistance. As mentioned previously in Section 3.1, a higher Nrresults in a shorter HRT, implying that the time for ozone gas–liquid mass transfer becomes shorter. Although a high Nr can decrease the liquid-film thickness and droplet size to improve the ozone gas–liquid mass transfer rate, however, the short contact time decreases the amount of ozone transferred.

3.4. Decomposition of DMP via catalytic ozonation

Fig. 6 presents the decomposition of DMP with and without Pt/␥-Al2O3 (Dash 220N) during the ozonation in HGRPB. For the case without catalyst, glass beads were used. The value ofDMPfor HG-Pt-OZ process reaches 92% at mA,in= 600 mg L−1(in 30 min). The decomposition of DMP is nearly completed at mA,in= 1200 mg L−1 (in 60 min). A comparison with HG-OZ indicates that DMP is decomposed rapidly because Dash 220N promotes the dissolved O3 to form highly reactive OH• radicals. The Pt on the Al2O3 sur-face is effective for the decomposition of O3in gaseous as well as aqueous phase[18,19]. The mechanism of the heterogeneous cat-alytic ozonation in water is shown inFig. 7which was explained by Legube and Vel Leitner[4]. Based on the mechanism, the Pt on the Dash 220N surface assists the generation of OH•radicals. The organic molecules (DMP and intermediates) are adsorbed on the Dash 220N surface and oxidized via an electron-transfer reaction. Then the organic radical species are desorbed from the Dash 220N and oxidized by OH•radical or O3.

Combination of HG-Pt-OZ with UV irradiation (HG-UV-Pt-OZ) can further enhance the decomposition of DMP. As shown in

Fig. 8. Mineralization extents (TOC) of different HG processes. C0, QG, mS, CAG,in, UV, VL, Nr, T: as specified inFig. 3. Reaction time = 1 h.

Fig. 6, after 10 min oxidation (mA,in= 200 mg L−1), DMP of HG-UV-Pt-OZ is 65%, while that of HG-Pt-OZ without UV irradiation is 48%. The valueDMPof HG-UV-Pt-OZ and HG-Pt-OZ at 30 min (mA,in= 600 mg L−1) are 94% and 92%. Both of these two catalytic ozonation processes (with Pt/␥-Al2O3or UV radiation) can com-pletely decompose the DMP at 50 min (mA,in= 1000 mg L−1).

3.5. Decomposition of TOC via various Higee processes

Fig. 8compares the values ofTOCof the HG related processes. After 1 h reaction, the HG-UV-OZ and HG-Pt-OZ have about the same decomposition efficiency of TOC (TOC) of 56% and 57%, respectively. The result showed that both the UV irradiation and Pt catalyst can enhance the decomposition of dissolved ozone to form OH•radicals, assisting the decomposition of TOC. Although ozone was fed and transferred in the HGRPB, the organics in solution were oxidized in HGRPB as well as in photolysis reactor with residual ozone. The transferred ozone is adsorbed on the Dash 220N surface and forms OH•radical to react with organics during the course of catalytic ozonation. The radicals in turn enhance the decomposition reactions of DMP and intermediates.

To further promote the decomposition ability of organic in the HG-UV-OZ and HG-Pt-OZ processes, the UV irradiation is combined with Pt/␥-Al2O3 catalyst in HG-OZ forming a novel HG-UV-Pt-OZ

(7)

process. Experiment result showed that, under the same ozone dosage, the HG-UV-Pt-OZ process has the highestTOCof 68%, com-paring with other processes examined in this study.

A simplified scheme of the decomposition pathways of DMP for the catalytic ozonation with UV is shown inFig. 9. Both the DMP and intermediates are attacked by O3and OH•radical. The initial attack of oxidants on DMP during ozonation is via the electrophilic addition to form aromatic acids. Then the aromatic rings are broken down by O3and OH•radical to form short-chain organic acids such as oxalic acid, glyoxalic acid and formic acid. These short-chain organic acids are further decomposed to CO2and H2O eventually. 4. Conclusions

The decomposition of DMP via HG-OZ, HG-UV-OZ, HG-Pt-OZ and HG-UV-Pt-OZ is investigated in this paper. Several conclusions can be drawn as follows.

(1) Packing Pt/␥-Al2O3catalyst (Dash 220N) in HG-OZ process can effectively enhance the decomposition of DMP under the exper-imental conditions of this study. TheTOCof DMP increases from 38% of HG-OZ to 57% of HG-Pt-OZ because of the formation of OH•radicals.

(2) The decomposition rate of DMP is dependent on the rotating speed Nrof packing bed. The thickness of liquid film and droplet size decrease with highly centrifugal force. As Nris higher than 1200 rpm, the effect of Nris greatly reduced.

(3) The use of UV irradiation in HG-Pt-OZ process signifi-cantly enhances both the decomposition and mineralization rate of DMP. The HG-UV-OZ process mainly absorbs UV-C ( = 200–280 nm) energy to decompose ozone to form high active radicals. The photolysis of DMP via UV irradiation (HG-UV) only is negligible.

(4) The combination of UV radiation in HGRPB system can enhance the decomposition of dissolved ozone to form OH• radicals, assisting the decomposition of TOC.

Acknowledgment

The authors would like to thank the National Science Council of Taiwan for the financial support of the work (NSC 94-2211-E-267-001).

References

[1] B.S. Oh, Y.J. Jung, Y.J. Oh, Y.S. Yoo, J.W. Kang, Application of O3, UV and O3/UV pro-cesses to reduce diethyl phthalate and its estrogenic activity, Sci. Total Environ. 367 (2006) 681–693.

[2] L.S. Li, W.P. Zhu, P.Y. Zhang, P. Lu, Q.Y. Zhang, Z.L. Zhang, UV/O3-BAC process for removing organic pollutants in secondary effluents, Desalination 207 (2007) 114–124.

[3] B. Langlais, D.A. Reckhow, Ozone in Water Treatment: Application and Engi-neering, Lewis Publishers, Chelsea, Michigan, 1991.

[4] B. Legube, N.K. Vel Leitner, Catalystic ozonation: a promising advanced oxidation technology for water treatment, Catal. Today 53 (1999) 61– 72.

[5] J. Lin, A. Kawai, T. Nakajima, Effective catalysts for decomposition of aqueous ozone, Appl. Catal. B 39 (2002) 157–165.

[6] C. Ramshaw, R.H. Mallinson, Mass Transfer Process, U.S. Patent 4,283,225 (1981).

[7] C.C. Lin, The Study of Higee, Ph.D. dissertation, Department of Chemical Engi-neering, National Taiwan University, Taipei, Taiwan, 1999.

[8] Y.H. Chen, C.Y. Chang, W.L. Su, C.C. Chen, C.Y. Chiu, Y.H. Yu, P.C. Chiang, S.I.M. Chiang, Modeling ozone contacting process in a rotating packed bed, Ind. Eng. Chem. Res. 43 (2004) 228–236.

[9] Y.H. Chen, C.Y. Chiu, C.Y. Chang, Y.H. Huang, Y.H. Yu, P.C. Chiang, J.L. Shie, C.S. Chiou, Modeling ozonation process with pollutant in a rotating packed bed, Ind. Eng. Chem. Res. 44 (2005) 21–29.

[10] K. Guo, F. Guo, Y.D. Feng, J.F. Chen, C. Zheng, N.C. Gardner, Synchronous visual and RTD study on liquid flow in rotating packed-bed contactor, Chem. Eng. Sci. 55 (2000) 1699–1706.

[11] Y.S. Chen, The Effect of Viscosity on Mass Transfer in a Rotating Packed Bed, Ph.D. dissertation, Department of Chemical Engineering, National Taiwan University, Taipei, Taiwan, 2004.

[12] Y.H. Chen, Y.H. Huang, R.H. Lin, C.Y. Chang, Production of biodiesel from soybean oil using a rotating packed bed, in: 98th AOCS (American Oil Chemists’ Society) Annual Meeting & Expo, QC, Canada, May 13, 2007.

[13] C.C. Lin, W.T. Liu, Ozone oxidation in a rotating packed bed, J. Chem. Technol. Biotechnol. 78 (2003) 138–141.

[14] Y.H. Chen, C.Y. Chang, W.L. Su, C.Y. Chiu, Y.H. Yu, P.C. Chiang, C.F. Chang, J.L. Shie, C.S. Chiou, S. Chiang, Ozonation of CI Reactive Black 5 using rotating packed bed and stirred tank reactor, J. Chem. Technol. Biotechnol. 80 (2005) 68– 75.

[15] J.L. Shie, C.Y. Chang, J.H. Chen, W.T. Tsai, Y.H. Chen, C.S. Chiou, C.F. Chang, Cat-alytic oxidation of naphthalene using a Pt/Al2O3catalyst, App. Catal. B 58 (2005) 289–297.

[16] L.H. Nowell, J. Hoign, Interaction of iron (II) and other transition metals with aqueous ozone, in: Proceedings of 8th Ozone World Congress, IOA, Zurich, Switzerland, 1988, pp. E80–E95.

[17] G.Y. Peyton, W.H. Glaze, Destruction pollutants in water with ozone in combi-nation with ultraviolet radiation. 3. Photolysis of aqueous ozone, Environ. Sci. Technol. 22 (1988) 761–767.

[18] M.C. Wu, N.A. Kellyb, Clean-air catalyst system for on-road applications. I. Eval-uation of potential catalysts, App. Catal. B 18 (1998) 79–91.

[19] K.C. Cho, K.C. Hwang, T. Sano, K. Takeuchi, S. Matsuzawa, Photocatalytic perfor-mance of Pt-loaded TiO2in the decomposition of gaseous ozone, J. Photochem. Photobiol., A 161 (2004) 155–161.

數據

Fig. 1. Schematic diagram of HGRPB system.
Fig. 3. Time variations of C/C 0 for the decomposition of DMP in HGRPB sys- sys-tems. Initial concentration of DMP (C DMP0 or C 0 ) = 100 mg L −1 , Q G = 1 L min −1 , Q LR = 462 mL min −1 , power of UV lamp = 16 W as noted in specification, V L = 1 L, N r =
Fig. 5. Effects of m A,in on the decomposition of DMP for the HG-OZ process at various N r
Fig. 7. Scheme illustrating the possible mechanism for ozonation process in the presence of metal-support (metal oxide) catalyst (based on)

參考文獻

相關文件

The academic achievement of math of high-grade elementary school students is significant related to their SES and the self-concept in math, but is non-related to their

This study has analyzed 13 bidding cases of different bidding mechanisms in high tech plant construction projects in the past, and hoping to establish a fair mechanism in

For the next nitrogen delivery system, In this study, the high-tech industry, nitrogen supply, for example, to explore in depth the relationship between

In order to know whether the sediment matrix would interfere with the measurement of total Hg after microwave digestion, the following two sets of calibration graphs were

In studying habits, 72.2 percent of the subjects attended cram schools before senior high school, and they went to cram schools two to four times a week.. After entering senior

This research tries to understand the current situation of supplementary education of junior high school in Taichung City and investigate the learning factors and

The main purpose of this study is to explore the status quo of the food quality and service quality for the quantity foodservice of the high-tech industry in Taiwan;

The purpose of this study was to investigate the current situation of multicultural literacy and intercultural sensitivity of junior high school teachers in Taichung